Secondary cells using a rechargeable hydrogen storage negative electrode are known in the art. These cells operate in a different manner than lead-acid, nickel-cadmium or other prior art battery systems. The hydrogen storage electrochemical cell utilizes a negative electrode that is capable of reversibly electrochemically storing hydrogen. In one exemplification the cell employs a positive electrode of nickel hydroxide material, although other positive electrode materials may be used. The negative and positive electrodes are spaced apart in an alkaline electrolyte, and may include a suitable separator, spacer, or membrane therebetween.
Upon application of an electrical current to the negative electrode, the negative electrode material (M) is charged by the absorption of hydrogen: EQU M+H.sub.2 O+e.sup.- --- M-H+OH.sup.- (Charge)
Upon discharge, the stored hydrogen is released to provide an electric current: EQU M-H+OH.sup.- --- M+H.sub.2 O+e.sup.- (Discharge)
The reactions are reversible.
The reactions that take place at the positive electrode are also reversible. For example, the reactions at a conventional nickel hydroxide positive electrode as utilized in a hydrogen rechargeable secondary cell or battery are: EQU Ni(OH).sub.2 +OH.sup.- --- NiOOH+H.sub.2 O+e.sup.- (Charge), and EQU NiOOH+H.sub.2 O+e.sup.- --- Ni(OH).sub.2 +OH.sup.- (Discharge).
A cell utilizing an electrochemically rechargeable hydrogen storage negative electrode offers important advantages over conventional secondary batteries. Rechargeable hydrogen storage negative electrodes offer significantly higher specific charge capacities (ampere hours per unit mass and ampere hours per unit volume) than do either lead negative electrodes or cadmium negative electrodes. As a result of the higher specific charge capacities a higher energy density (in watt hours per unit mass or watt hours per unit volume) is possible with hydrogen storage batteries than with the prior art conventional systems, making hydrogen storage cells particularly suitable for many commercial applications.
Suitable active materials for the negative electrode are disclosed in commonly assigned U.S. Pat. No. 4,551,400 to Sapru, Hong, Fetcenko and Venkatesan for HYDROGEN STORAGE MATERIALS AND METHODS OF SIZING AND PREPARING THE SAME FOR ELECTROCHEMICAL APPLICATION incorporated herein by reference. The materials described therein store hydrogen by reversibly forming hydrides. The materials of Sapru, et al have compositions of: EQU (TiV.sub.2-x Ni.sub.x).sub.1-y M.sub.y
where 0.2.times.1.0, 0 y 0.2 and M=Al or Zr; EQU Ti.sub.2-x Zr.sub.x V.sub.4-y Ni.sub.y
where 0.times.1.5, 0.6 y 3.5; and EQU Ti.sub.1-x Cr.sub.x V.sub.2-y Ni.sub.y
where 0.times.0.75, 0.2 y 1.0.
Reference may be made to U.S. Pat. No. 4,551,400 for further descriptions of these materials and for methods of making them.
Other suitable materials for the negative electrode are disclosed in commonly assigned U.S. Pat. No. 4,728,586 issued Mar. 1, 1988 in the names of Srinivasen Venkatesan, Benjamin Reichman, and Michael A. Fetcenko for ENHANCED CHARGE RETENTION ELECTROCHEMlCAL HYDROGEN STORAGE ALLOYS AND AN ENHANCED CHARGE RETENTION ELECTROCHEMICAL CELL, incorporated herein by reference. As described in the above referenced application of Venkatesan, et al, one class of particularly desirable hydrogen storage alloys comprises titanium, vanadium, nickel, and at least one metal chosen from the group consisting of aluminum, zirconium, and chromium. The preferred alloys described in Venkatesan, et al are alloys of titanium, vanadium, nickel, zirconium, and chromium, especially alloys having the composition represented by the formula: EQU (Ti.sub.2-x Zr.sub.x V.sub.4-y Ni.sub.y).sub.1-z Cr.sub.z
where x is between 0.0 and 1.5, y is between 0.6 and 3.5, and z is an effective amount less than 0.20.
The hydrogen storage alloy is formed from a melt. The production of hydrogen storage negative electrodes utilizing the preferred materials is difficult because these preferred materials are not only not ductile, but are in fact, of relatively great or high hardness. Indeed, these alloys can typically exhibit Rockwell "C" (R.sub.C) hardnesses of 45 to 60 or more. Moreover, in order to attain high surface areas per unit volume and per unit mass, the alloy must be in the form of small ash or flake-like particles. In a preferred exemplification, the hydrogen storage alloy powder must pass through a 200 U.S. mesh screen, and thus be smaller than 75 microns in size (200 U.S. mesh screen has interstices of about 75 microns). Therefore, the resulting hydrogen storage alloy material must be comminuted, e.g., crushed ground, milled, or the like, before the hydrogen storage material is fabricated into an electrode.
Comminution of bulk ingots of metal hydride, hydrogen storage alloy material is made more difficult by the fact that the materials described hereinabove are quite brittle, and therefore do not easily break into particles of uniform size and shape. In commonly assigned, copending U.S. patent application Ser. No. 247,569 filed Sept. 22, 1988 the inventors disclosed an economical, safe, hydride-dehydride cycle comminution process for the initial size reduction of bulk ingots of hydrogen storage alloy material to flakes of about 80-100 mesh size. While this process is quite excellent for initial size reduction of hydrogen storage alloy, it is inadequate for the task of further comminuting particulate metal hydride, hydrogen storage alloy powder to particles of 75 microns or less (i.e., 200 mesh or less). The approximately 200-400 mesh size distribution has been experimentally determined to provide performance which is superior to other sizes of material in the negative electrode of metal hydride, hydrogen storage electrochemical cells.
Generally, it can be said that any method which can accomplish the objective of economical size reduction of the metal hydride material is a potential candidate. However, there are numerous characteristics of the material which require special handling instrumentation and precautions. These include: (1) inherent alloy powder hardness, i.e., approximately Rockwell "C" 60 hardness, consequently conventional size reduction processes of shear, abrasion and some kinds of impact such as ball mills, hammer mills, shredders, fluid energy, disk attrition are not very effective; (2) sensitivity to oxidation, such that comminution thereof must be done under an inert environment; this is for the dual reasons of providing a safe environment and maintaining acceptable electrochemical performance; (3) due to the crystal structure necessary for electrochemical activity, the microstructure of the material cannot be altered during grinding in an adverse manner, i.e., such as by atomization to produce powders directly from a melt; and (4) the particle size distribution; the size distribution must be a wide distribution with a maximum size of 75 microns (200 mesh), to provide superior electrochemical performance.
There are numerous methods for preparing metal powders. Since the alloys under consideration are at one stage molten, it might be considered that ultrasonic agitation or centrifugal atomization could be utilized on the liquid stream to prepare powders directly. This approach would not provide the desired alloy microstructure because the cooling rates are extremely fast. The atomization process also tends to provide a narrow distribution of particle size, which for this application is undesirable. Particle shape is also not optimal. Finally, it is difficult to provide a completely inert atmosphere, such that surface layers which are undesirable from an electrochemical outlook may be formed.
Another approach involves mechanical means. With conventional materials, ingots can first be crushed in a jaw crusher, followed by successively finer treatments until the desired particle size is achieved. As stated earlier, hydrogen storage materials are not amenable to these types of size reduction techniques due to the materials high hardness. Frequently, significant wear of the grinding medium is observed. A direct consequence of grinding media wear is the introduction of contaminants into the final powder. Iron and steel are well known contaminants which materially degrade performance in electrochemical cells of the type being considered.
Even if such mechanical means are contrived, it is difficult to attain the desired wide range of powder size since the materials are difficult to grind. A high fraction of powder just under the specified maximum value is frequently observed.
A further problem with conventional mechanical grinding means is observed during both unloading and cleaning of such equipment. The powder has a high surface area. In fact, the surface area of the comminuted material can render it pyrophoric. Thus, a hazardous condition exists for either fire or explosion if the comminuted material, oxygen, and an ignition source, such as a spark, should coincide. Thus, the grinding system must be relatively easy to clean and should be designed in such a way that residual powder can be minimized or eliminated. Thus, for example, although a high energy ball mill does in fact accomplish excellent size reduction, it is not practical, as powder becomes imbedded into the media, presenting a significant safety hazard during routine maintenance.
Also, grinding processes invariably have low yield factors, that is, an unacceptably high fraction of the feed stock of particulate metal hydride hydrogen storage alloy is oversized. After comminution, the material invariably requires classification to separate that material which is oversize. The process should be integrated such that the oversize powder is separated from the material within specification and accumulated to be rerun through the grinding system.
The comminution process should provide a wide distribution of particle size under the prescribed maximum. This distribution requires careful classification. An air classifier, or cyclone, can be considered. Problems occur because a cyclone tends to provide a narrow powder distribution, must be calibrated, and is subject to operating condition fluctuations. Further, a cyclone tends to use large quantities of gas as a carrier. Since inert gas must be used, it is potentially very costly. Consequently, other effective, cost efficient classification means must be considered.
For hydrogen storage materials used in electrochemical cells, it is desirable to have the powder as oxide free as possible. Consequently, all processing must be done in an inert environment. Generally, it can be said that the higher the temperature, the more sensitive the powder is to oxidation. Thus, it is desired that the integrated grinding/classification process be run in an inert environment, at the lowest temperature possible. If the temperature can be low enough for nitrogen to be used as the inert gas, there are substantial cost savings to be realized compared to argon gas.
In principle, fluid energy pulverizers have characteristics making this type of processing possible. The basic principle is to use the material to impact on itself for size reduction. Conventionally, the materials are contained in a gas stream which is under pressure, causing high speed impact of the material. From practical experience, this type of system did not work well. Size reduction was not accomplished with a high yield, large amounts of gas were required to accelerate the material and the equipment was not reliable.
It can thus be seen that heretofore the step of providing appropriately sized metal hydride, hydrogen storage alloy material has been a difficult one in the overall fabrication process of the metal hydride negative electrode.
One apparatus which may be adapted, with considerable reengineering of system operation parameter, to size reduce particulate metal hydride material, is shown in U.S. Pat. No. 4,641,787 to Peterson, et al. While the Peterson, et al comminution system is generally similar to that disclosed herein, providing powdered metal hydride hydrogen storage alloy material for subsequent fabrication of electrochemically rechargeable negative electrodes is not taught, suggested or disclosed therein. Further, the is no suggestion, teaching nor disclosure of using such a system to provide one step in an integrated fabrication process such as that disclosed hereinbelow.
Thereafter, the comminuted hydrogen storage alloy powder is applied to a suitable wire mesh or wire screen current collector to form a negative electrode. Various methods of manufacturing electrodes of strip configuration have been previously proposed and utilized, e.g. for cadmium negative electrodes. However, these methods and their associated equipment cannot be used with the high hardness, flake or ash-like active powdered hydrogen storage materials contemplated herein. For example, a system for making cadmium battery plates is suggested in U.S. Pat. Nos. 3,894,886 and 3,951,688. The system described therein utilizes an electrochemically active thioxotropic paste to carry the active material and is thus inapplicable to the herein contemplated and described negative electrode materials.
Another method of making silver strip electrodes involves feeding a free-flowing powder onto a moving paper carrier web. Vibrating doctor blades spread the powder on to the carrier to a pre-determined thickness. A grid structure or mesh is introduced to the powder and carrier. A single rolling mill compresses the grid and powder on the carrier, and thereafter the carrier is withdrawn. The remaining web of silver electrode material impregnated into a grid is then sintered. After the sintered silver electrode web leaves the sintering furnace, it is cut into strips for use in silver-zinc electrochemical cells.
The above described system cannot be used with the high hardness, powdered hydrogen storage alloy material herein, because the hydrogen storage alloy powder does not behave in the same way as the silver powder of the silver electrode production line. When the high hardness metal hydride powder used herein is compressed onto a paper carrier, the powder particles stick to and become embedded in the paper. Web tearing or other web damage can result. In addition, the hydrogen storage alloy electrode materials are typically deposited as a relatively thin layer of flake-like particles on a smooth, hard carrier. It has been found that doctor blades are ill-suited to provide a precisely controlled thickness or depth of powder, because the powder flakes or ash-like particles tend to agglomerate and build up in front of the blades. A powder layer of irregular thickness and density with regions of inadequate depth results. Uniformity of powder depth, and consequently uniformity of the amount of active material per unit area, is necessary to provide a uniform electrode strip. A uniform strip thickness is essential for battery electrodes if the battery is to have a uniform current density, and make efficient use of space within the cell.
Hydrogen storage electrodes for sealed cells have previously been prepared by various methods as described in, for example, U.S. Pat. No. 4,670,214 to Douglas Magnuson, Merle Wolff, Sam Lev, Kenneth Jeffries, and Scott Mapes for "METHOD FOR MAKING ELECTRODE MATERIAL FROM HIGH HARDNESS ACTIVE MATERIALS", the disclosure of which is incorporated herein by reference. The method disclosed therein is not however, altogether adequate for the most efficient method of continuously producing large area hydrogen storage alloy negative electrodes. Specifically, the previous method allowed the active negative electrode powder to be transported through an ambient environment to the fabrication apparatus on a temporary web, substrate or carrier means. Exposure to the atmosphere resulted in the oxidation of catalytically active sites in the active material. Additionally, the use of a strip of non-reusable, organic polymeric carrier material, such as, for example Mylar (a registered trademark of Dupont), to transport the active material to the mesh substrate and through the compaction process adversely affected electrode production in two critical ways: cost of manufacture and accuracy of production. Specifically, the carrier, since it was not reusable, added cost to the manufacture of the negative electrode. More importantly, the carrier had a tendency to deform in a non-uniform manner when subjected to roller mill compaction. This resulted in non-uniformities in the thickness of the electrode web after passing through the roller mills.
A further problem often encountered in the fabrication of hydrogen storage negative electrodes are the deleterious effects of oxygen and water. While alluded to above with respect to the transfer of the material to the compaction process, the problem is particularly acute during, and immediately after sintering. Oxygen and water cause the electrode material to be less functional by bonding to catalytic sites, thereby requiring more charge/discharge cycles to activate these sites. Additionally, oxygen and water react with the electrode material in the sintering process, lowering the ultimate electrode capacity. This necessitated purging the electrode web of oxygen and water. The preferred prior art method called for heat treating the electrode web in a large volume argon purged environment. This process of course required a significant economic investment in argon. The prior art also failed to make use of a controllable system for the incorporation of hydrogen in the electrode web, which hydrogen sets the charge state, i.e., partially charges the electrode web.
Finally, prior art continuous systems were subject to frequent mechanical failure as a result of the electrode web "walking". "Walking" is the lateral movement of the electrode web across the roller mills. This walking results in numerous tears and consequent splices in the electrode web, which is costly in terms of lost product, and is incompatible with apparatus used in subsequent downstream processing steps.
Accordingly, there exists a need for an improved method of fabricating metal hydride, hydrogen storage alloy negative electrodes, including steps directed towards providing appropriately sized hydrogen storage alloy powder.